Year: 2019
Type: conference-paper
Abstract: Rainfall prediction is one of the most important and at the same time challenging task. Meteorologists can predict weather patterns such as rainfall based on atmospheric parameters such as Humidity, Temperature, etc. This paper presents research on rainfall prediction based on historical dataset through neural network by training a network and testing it. Mean squared error (MSE) is used to generalize the performance of the model. Three different dataset, training algorithm and hidden layer setting is used for prediction. The results obtained reveals that ideally all three training algorithm is producing good results as MSE is closer to zero. Undoubtedly, neural network proves to be most appropriate technique for forecasting various weather phenomena such as rainfall. It can be further alluded that Bayesian Regularization tends to give lower MSE values compare to the other two training algorithms.
Authors: Ashneel Chand, Ravneil Nand
Institution: University of the South Pacific
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
SDG: Climate action
Open Access status: bronze
OpenAlex ID: W3047765451